Method and device for detecting road partitions

Abstract

A detection method and a detection device for road segmentation objects are provided. Obtain the disparity map of the road area and the road vanishing point; perform coordinate transformation on the disparity point in the disparity map based on the road vanishing point; and detect the road segmentation object based on the coordinate transformed disparity point. The detection method and device of the road segmentation object of the present invention can perform coordinate transformation on the disparity point in the original disparity map based on the road vanishing point, so as to detect the road segmentation object based on the transformed disparity point.

Description

technical field [0001] The present invention relates to the field of object detection, and more particularly to a method and device for detecting road segment objects. Background technique [0002] The application of driver assistance systems is becoming more and more popular. The road or lane warning system (Lane / Road detection warning, LDW / RDW) is a subsystem of the driving assistance system, which can avoid collisions and determine the driving direction more accurately. Road or lane detection is very critical for LDW / RDW systems, and further processing, such as warning, is possible only on the basis of knowing the road information. Generally, roads or lanes are detected by detecting road dividers. [0003] Road dividers include lane lines, shoulder stones, fences, and other objects that can identify road areas and lanes, and lane lines can include white lines, yellow lines, and so on. [0004] In existing methods, the disparity map can be used to detect lane lines. Fo...

AI Technical Summary

Problems solved by technology

However, in some cases, such as figure 1 As shown in the oval box in , there are very few disparity pixels representing road boundaries or road divisions on the sparse disparity map, so it may be difficult to directly perform road detection on the disparity map

Moreover, in actual road conditions, roads are not always straight, and various curves or detours are also common. However, most of the current detection methods are for straight roads, and usually cannot deal with non-straight roads, such as figure 1 detour shown in

Images